A GIS-based Atmospheric Dispersion Modeling Project for Introductory Air Pollution Courses

摘要

Students enrolled in introductory air pollution courses can have difficulty understanding or visualizing dispersion modeling using the Gaussian plume equation. They can also be challenged by the changing nature of the plume as it travels downwind and combines with other plumes within a given area. Calculations by hand or in a spreadsheet generally focus on manipulating one or two variables and may only plot one plume in one dimension. To address such limitations, several years ago we developed a customized application integrating a Geospatial Information Science (GIS) program, specifically ESRI's ArcMap 9.1, with a Matlab script. When used together with specified atmospheric and source parameters for a Gaussian plume, these programs enabled the graphical display of a grid of downwind concentrations on a map. Recently we conducted a comprehensive redesign of the project using only ESRI's ArcGIS 10.0 for both concentration calculations and plotting. The project scenario asks teams of approximately four students, who comprise a "company", to locate several new cement factories and power plants within a given city, calculate the pollutant uncontrolled emissions rate, and identify mitigation techniques (e.g., increased stack height or incorporation of pollution control devices) to meet the US National Ambient Air Quality Standards (NAAQS). Using a custom interface in ArcGIS 10.0, students vary atmospheric stability conditions, stack heights, wind speed, and calculated controlled emission rates to create an array of downwind plume concentrations from all existing and new sources, which are plotted on a city map. Since costs increase for higher stacks and more effective control devices, students attempt to locate sources in a manner that will minimize costs. In ArcGIS 10.0, multiple plume concentrations are then summed and the resulting impacts on four major urban categories (residential, schools, religious complexes, and hospitals) are quantified and depicted. The student company with the most optimized solution (i.e., lowest total cost) that meets the NAAQS for PM_(10), the chosen pollutant, under given atmospheric conditions is awarded the bid. While the application creates a relatively simple model of the dispersion process, it helps students visualize dispersion on a macro-scale, and the specific effect of the variation of each parameter on downwind concentrations. Post-project assessment data indicates that all students (n=10) consider themselves knowledgeable on how to use the Gaussian dispersion model to solve for downwind pollutant concentrations. Additionally, 80% of students surveyed post-project indicated that the dispersion project increased their knowledge of Gaussian dispersion modeling for air pollutants. Students also reported that this project increased their familiarity with ArcGIS and that the project is a useful interdisciplinary coupling of environmental engineering and GIS.